Project Summary Dopamine (DA) is an established neuromodulator in brain reward pathways. It is increasingly recognized that regulatory factors acting within the striatum sculpt local DA release, complementing the role of midbrain DA neuron activity in supplying DA to target regions. We have reported that glucose-induced increases in levels of the metabolic hormone insulin acts as a reward signal in the nucleus accumbens (NAc). Using ex vivo striatal slices, we discovered that insulin enhances DA release in the NAc and dorsal striatum, and that insulin responsiveness is lost in rodents that are obese from a high-fat high-sugar (HF-HS) diet. Our companion behavioral studies show that insulin action in the NAc is necessary for flavor-nutrient learning with glucose- containing solutions, and for the escalation of glucose intake seen during training. Although understanding of this reward pathway is far from complete, we have established the pivotal role of a striatal microcircuit involving DA axons and striatal cholinergic interneurons (ChIs), both of which express insulin receptors (InsRs). An abundant literature indicates that acetylcholine (ACh) from ChIs promotes DA release via ?2-subunit containing nicotinic ACh receptors (?2-nAChRs) on DA axons. We showed previously that insulin acting at InsRs increases ChI excitability, and that the DA-boosting effect of insulin is prevented by ?2-nAChR-selective antagonism, and is absent in mice that lack striatal ACh synthesis (ChAT KO mice). However, key elements required to understand and harness this pathway are missing, including: 1) cellular mechanisms by which insulin increases ChI activity; 2) specific components of flavor-nutrient learning influenced by insulin and whether they are blocked by ?2- nAChR antagonists and impaired by a HF-HS diet; and 3) patterns of insulin-dependent synaptic plasticity in NAc medium spiny output neurons (MSNs) that underlie flavor-nutrient learning. These missing elements will be addressed in three specific aims that capitalize on the complementary expertise of the PIs. Overall, our previous and pilot data show that NAc insulin signaling is necessary for flavor-nutrient learning, which guides food choice and consumption based on predicted nutritive yield. This insulin-dependent reward learning is impaired in subjects with InsR subsensitivity induced by HF-HS feeding, leading to maladaptive consumption. By identifying cellular mechanisms and pathway-specific plasticity that drive nutritive learning, this project will not only answer key questions about NAc insulin, but also indicate targets that might bypass subsensitive NAc InsRs and restore healthy eating.